: Pompe disease results from mutations in the gene for acid ?-glucosidase (GAA) ? an enzyme necessary to degrade lysosomal glycogen. Early-onset disease occurs in the absence of functional GAA which leads to cardiorespiratory failure early in life. Late-onset disease is associated with reduced GAA activity and gradual progression to respiratory failure. Work from our first two grant cycles indicates neural involvement in respiratory failure in Gaa-/- mice and Pompe patients. This is relevant since the standard of care ? intravenous enzyme therapy using recombinant GAA - does not reach the central nervous system (CNS) and patients still progress to respiratory failure. Our overarching hypothesis is that adeno-associated virus (AAV) therapy is capable of restoring life-long GAA expression throughout the CNS, skeletal and cardiac muscle, thereby preserving cardiorespiratory function and prolonging life. Aim 1 focuses on AAV therapy for early-onset disease which requires early life treatments that can prevent both respiratory and cardiac failure. To better study this problem, we created a Gaa null (Gaa-/-) rat model which recapitulates the early onset phenotype with cardiorespiratory pathology and early mortality. Preliminary data indicate that neonatal AAV-GAA therapy (desmin promoter, AAV9 serotype) evokes no detectable immune response, mitigates cardiac and respiratory pathology and prevents early death. Thus, we hypothesize that a single intravenous AAV-GAA dose in young rats can drive persistent and widespread GAA expression and extend the Pompe rat lifespan. Aim 2 addresses late onset Pompe disease in which respiratory failure is the primary cause of mortality. Based on data from our first two grant cycles we hypothesize that neural directed AAV-GAA therapy in adult Pompe rats is sufficient to prevent respiratory decline and extend the lifespan. By packaging AAV-GAA with muscle (creatine kinase 8), neural (synapsin) or tissue specific (muscle and neural, desmin) promoters, and delivering the vector intrathecally, intravenously, or both, we can drive GAA expression in a manner that will determine if neural correction is necessary and sufficient to prevent decline. The aforementioned Gaa null rat will be used to test proof-of-concept for neural vs. muscle correction in the absence of endogenous GAA activity. We will also use another new Pompe rat model in which CRISPR/cas9 has been used to insert the most common human gene mutation causing late-onset Pompe disease (IVS1) into the rat genome. This is important because the IVS1 mutation leads to low but not absent GAA activity and is associated with delayed progression to respiratory failure. The proposed work is significant because current therapeutic strategies in Pompe disease only delay disease progression with eventual respiratory failure. The strategies proposed here will also contribute to the broader goal of advancing gene therapy for neurodegenerative conditions and autosomal recessive diseases.